hurali97 · Oct 14, 2026 · Oct 14, 2026 · Oct 14, 2026 · Oct 14, 2026 · Oct 15, 2026
diff --git a/README.md b/README.md
@@ -22,6 +22,9 @@ Install a built app, launch or stop it, and open a deep link or URL to jump to a
 ### Device settings and network conditions
 Change the running device's appearance, text size, rotation, location, and status bar without restarting its stream. Switch the network to LTE, 3G, Edge, or Offline to see how the app behaves on a slow or missing connection, then back to Normal. Available settings depend on the platform and device kind.
 
+### App Flow
+Build a visual map of your app's screens and the navigation between them, for Expo Router and React Navigation projects. **Structure only** reads the navigation out of your source in seconds and needs no device; **Screen UI + Structure** also walks the app on an iOS simulator and photographs every screen it reaches, optionally opening each screen's sheets, modals, and dialogs. The walk runs your project's own `ios` script and reaches each screen through in-app navigation, a deep link, or a path you recorded; record one by hand on the live wireframe for a screen it cannot reach. Explore with AI lists the screens the deterministic pass missed and retries only the ones you pick. Maps, captures, and recorded paths are stored per project under `~/Library/Application Support/mobile-dev/app-flow`.
+
 ### Native and JavaScript logs
 Read iOS unified logs, Android logcat, and Metro console messages together. Follow the foreground app, search and filter logs, and send an error and its stack trace to chat.
 

diff --git a/apps/claude/.claude-plugin/plugin.json b/apps/claude/.claude-plugin/plugin.json
@@ -1,7 +1,7 @@
 {
   "name": "mobile-dev",
   "displayName": "Mobile Dev",
-  "description": "Everything you need for mobile development with your agent. Live devices (simulators, emulators, and physical devices) beside your chat, app and device control, screen annotations, logs, app storage inspection, and dedicated tools to profile your app's performance (CPU, memory, FPS, CPU and React profiles with TraceSift). Requires an Apple Silicon Mac.",
+  "description": "Everything you need for mobile development with your agent. Live devices (simulators, emulators, and physical devices) beside your chat, app and device control, screen annotations, logs, app storage inspection, visual app flow maps of your screens and navigation, and dedicated tools to profile your app's performance (CPU, memory, FPS, CPU and React profiles with TraceSift). Requires an Apple Silicon Mac.",
   "author": {
     "name": "Callstack",
     "email": "hello@callstack.com"

diff --git a/apps/claude/README.md b/apps/claude/README.md
@@ -15,6 +15,9 @@ SwiftUI, and other native mobile projects.
   [agent-device](https://agent-device.dev/) runtime.
 - **App control:** install, launch, and stop apps, open deep links, send test pushes,
   and set permissions and languages.
+- **App Flow:** a visual map of your app's screens and the navigation between them,
+  read from your Expo Router or React Navigation source and optionally captured by
+  walking the app on an iOS simulator.
 - **Logs:** iOS unified logs, Android logcat, physical iPhone logs, and Metro console
   messages.
 - **Performance:** CPU, memory, and FPS recordings shown as interactive charts in

diff --git a/apps/claude/manifest.json b/apps/claude/manifest.json
@@ -2,7 +2,7 @@
   "manifest_version": "0.3",
   "name": "mobile-dev",
   "display_name": "Mobile Dev",
-  "description": "Everything you need for mobile development with your agent. Live devices (simulators, emulators, and physical devices) beside your chat, app and device control, screen annotations, logs, app storage inspection, and dedicated tools to profile your app's performance (CPU, memory, FPS, CPU and React profiles with TraceSift). Requires an Apple Silicon Mac.",
+  "description": "Everything you need for mobile development with your agent. Live devices (simulators, emulators, and physical devices) beside your chat, app and device control, screen annotations, logs, app storage inspection, visual app flow maps of your screens and navigation, and dedicated tools to profile your app's performance (CPU, memory, FPS, CPU and React profiles with TraceSift). Requires an Apple Silicon Mac.",
   "author": {
     "name": "Callstack"
   },

diff --git a/apps/codex-plugin/.codex-plugin/plugin.json b/apps/codex-plugin/.codex-plugin/plugin.json
@@ -1,6 +1,6 @@
 {
   "name": "mobile-dev",
-  "description": "Keep a simulator beside the chat while building, running, or debugging iOS and Android apps. Stream devices, read native and Metro logs, analyze CPU and React profiles with TraceSift, render interactive CPU, memory, and FPS recordings, and control apps with Mobile Dev tools.",
+  "description": "Keep a simulator beside the chat while building, running, or debugging iOS and Android apps. Stream devices, read native and Metro logs, analyze CPU and React profiles with TraceSift, render interactive CPU, memory, and FPS recordings, map your app's screens and navigation with App Flow, and control apps with Mobile Dev tools.",
   "author": {
     "name": "Callstack",
     "email": "hello@callstack.com"
@@ -75,7 +75,7 @@
   "interface": {
     "displayName": "Mobile Dev",
     "shortDescription": "Devices, logs, and performance",
-    "longDescription": "Keep an iOS simulator or Android device beside your chat while building, running, or debugging Expo, React Native, SwiftUI, and other mobile apps. Monitor native CPU, memory, and device-wide Display FPS, analyze local CPU and React profiles with TraceSift cards, render interactive performance recordings in chat, read native and Metro logs, and inspect and control apps with Mobile Dev tools.",
+    "longDescription": "Keep an iOS simulator or Android device beside your chat while building, running, or debugging Expo, React Native, SwiftUI, and other mobile apps. Monitor native CPU, memory, and device-wide Display FPS, analyze local CPU and React profiles with TraceSift cards, render interactive performance recordings in chat, read native and Metro logs, map your app's screens and navigation with App Flow, and inspect and control apps with Mobile Dev tools.",
     "developerName": "Callstack",
     "websiteURL": "https://github.com/callstackincubator/mobile-dev",
     "supportURL": "https://github.com/callstackincubator/mobile-dev/blob/main/SUPPORT.md",

diff --git a/apps/cursor/.cursor-plugin/plugin.json b/apps/cursor/.cursor-plugin/plugin.json
@@ -1,6 +1,6 @@
 {
   "name": "mobile-dev",
-  "description": "Stream iOS and Android devices into the chat while building, running, or debugging mobile apps. Read native and Metro logs, inspect and control apps, and record CPU, memory, and FPS. Apple Silicon development preview.",
+  "description": "Stream iOS and Android devices into the chat while building, running, or debugging mobile apps. Read native and Metro logs, inspect and control apps, map your app's screens and navigation with App Flow, and record CPU, memory, and FPS. Apple Silicon development preview.",
   "author": {
     "name": "Callstack",
     "email": "hello@callstack.com"

diff --git a/docs/mcp-tools.md b/docs/mcp-tools.md
@@ -39,6 +39,25 @@
 | `mobile_read_display_fps` | Read FPS intervals and connection status |
 | `mobile_read_performance_frames` | Read Android jank/pacing statistics and page through exact display frames |
 | `mobile_display_fps_close` | Stop FPS collection and release the tracing connection |
+| `mobile_appflow_start` | Start a non-blocking structure build or screen sweep; Adaptive (the default) tries Runtime → Deep Link → verified replay without AI, while `ai` continues unresolved screens with interactive exploration. Legacy Fast, Auto and Thorough modes remain accepted. |
+| `mobile_appflow_status` | Read current or persisted App Flow status |
+| `mobile_appflow_result` | Read the current or persisted screen graph |
+| `mobile_appflow_events` | Poll buffered progress and manual-recording events past a sequence cursor (app only) |
+| `mobile_appflow_pause` | Pause a running sweep at a safe boundary (app only) |
+| `mobile_appflow_cancel` | Cancel a run and any parked agent decision |
+| `mobile_appflow_launch_app` | Launch the selected project without starting a sweep (app only) |
+| `mobile_appflow_resolve_prompt` | Resolve a human-action gate after the user acts (app only) |
+| `mobile_appflow_projects` | List the project folders App Flow can open (app only) |
+| `mobile_appflow_choose_project` | Select a listed project or open the system folder picker (app only) |
+| `mobile_appflow_manual_start` | Start recording a path to one screen (app only) |
+| `mobile_appflow_manual_act` | Perform and record one manual path action (app only) |
+| `mobile_appflow_manual_stop` | Save the active manual path recording (app only) |
+| `mobile_appflow_manual_cancel` | Discard the active manual path recording (app only) |
+| `mobile_appflow_manual_replay` | Replay a saved path to one screen (app only) |
+| `mobile_appflow_runtime_replay` | Navigate to one screen through the running app's navigator, no recording needed (app only) |
+| `mobile_appflow_next_decision` | Long-poll for the next active-agent sweep decision (model only) |
+| `mobile_appflow_submit_decision` | Submit one role-valid JSON decision and wait briefly for the next (model only) |
+| `mobile_appflow_screenshot` | Read an authorized App Flow PNG (app only) |
 
 The plugin servers for Codex, Claude and Cursor add the pinned agent-device runtime's commands as `device_*` tools with compact, validated schemas, including `device_open`, `device_snapshot`, `device_screenshot`, `device_press`, `device_fill`, `device_type`, `device_scroll`, `device_wait`, `device_find`, `device_get`, `device_is`, `device_close`, and debugging tools. The agent reads screens and sends input with them; `device_screenshot` also returns the saved image. Their annotations let hosts run reads, device input and app control without asking, as with the panel's tools, while installs, scripts, app-defined events, file outputs, the clipboard and development servers still ask (see [Architecture](architecture.md#runtime)). Their input schemas describe each command; `logs` and `perf` are left out for Mobile Dev's own tools. The agent-device skill explains the control loop and ships the runtime's own workflow guide.